ALBERT

Description: The choice of planting density is a primary silvicultural decision in plantation management which considers the trade-off between individual tree size and total stand production, affecting the type, quantity and quality of products throughout the rotation. Trends in size and production with planting density are generally well known, however, less so is the interacting effect of site quality. Consequently, a case study in which basal area and basal area growth of Eucalyptus globulus Labill. plantations on five site qualities (122–435 m3·ha−1) planted at six densities (625 trees·ha−1, 4 m × 4 m; 833 trees·ha−1, 3 m × 4 m; 1000 trees·ha−1, 4 m × 2.5 m; 1250 trees·ha−1, 4 m × 2 m; 1667 trees·ha−1, 3 m × 2 m; and 2000 trees·ha−1, 3 m × 1.75 m) were used to investigate this interaction. As expected, both mean tree diameter of the whole stand and the basal area of the largest diameter 200 trees·ha−1 (D200 trees) were higher at lower planting densities, whereas whole stand basal area was greater at higher planting densities. However, there were no significant (P 〉 0.32) interactions between planting density and site quality for D200 or stand basal area, which contrasts with thinning responses in similar stands. This simplifies management considerations and suggests that trials at a given site quality may provide useful information about responses to planting density at other site qualities for the studied species.

Description: Fire scars are widely used to reconstruct fire history, yet patterns of scarring are poorly understood, hampering effective sampling and analysis. Factors that influence the probability a tree will receive a scar (SP) and the fraction of trees that scar (SF) are little studied. We analyzed scarring in 16 fires in ponderosa pine (Pinus ponderosa Douglas ex P. Lawson & C. Lawson) forests in northern Arizona. SP was significantly related to char height, presence of a preceding scar, tree diameter, and years since a preceding fire. Mean SF was 0.375, but varied from 0.121 to 0.728, with SF significantly higher with higher mean char height, larger scar dimensions, higher fire severity, larger tree diameter, and where no preceding fire had burned within 30 years. The expected healing times exceeded 55 years for 33% of scars and 100 years for 11% of scars. Scars with a preceding scar were 38% larger than new scars, with expected healing about 20–25 years longer. Scars were clustered, particularly at scales from 〉20 to 〉40 m. Scar directions generally aligned with fire-spread directions, which were complex. Variability in SF complicates fire-history methods that use fire counts rather then area burned. Methods that account for spatial and temporal variability in the abundance of evidence are needed.

Description: Helmers, M. J., Zhou, X., Baker, J. L., Melvin, S. W. and Lemke, D. W. 2012. Nitrogen loss on tile-drained Mollisols as affected by nitrogen application rate under continuous corn and corn-soybean rotation systems. Can. J. Soil Sci. 92: 493–499. Nitrate-nitrogen (NO3-N) loss from production agricultural systems through subsurface drainage networks is of local and regional concern throughout the Midwestern United States. The increased corn acreage and the practice of growing continuous corn instead of a corn-soybean rotation system due to the increasing demand for food and energy have raised questions about the environmental impacts of this shift in cropping systems. The objective of this 4-yr (1990–1993) study was to evaluate the effect of nitrogen (N) application rate (0–168 kg N ha−1 for corn following soybean and 0–224 kg N ha−1 for corn following corn) on NO3-N concentration, NO3-N losses, and crop yields in continuous corn and corn-soybean production systems on tile-drained Mollisols in north central Iowa. The results show that NO3-N concentrations from the continuous corn system were similar to NO3-N concentrations from the corn-soybean rotation at equivalent N application rates.When extra N fertilizer (approximately 56 kg N ha−1) was applied to continuous corn than the corn-soybean rotation, this resulted in 14–36% greater NO3-N concentrations in subsurface drainage from the continuous corn system. While corn yield increased as N application rate increased, corn yields at the recommended N application rates (112–168 kg N ha−1) in the corn-soybean rotation were up to 3145 kg ha−1 greater than corn yields at the recommended application rates (168–224 kg N ha−1) in the continuous corn system. The corn-soybean rotation with recommended N application rates (168–224 kg N ha−1) appeared to be beneficial environmentally and economically.

Description: Cade-Menun, B. J., Bell, G., Baker-Ismail, S., Fouli, Y., Hodder, K., McMartin, D. W., Perez-Valdivia, C. and Wu, K. 2013. Nutrient loss from Saskatchewan cropland and pasture in spring snowmelt runoff. Can. J. Soil Sci. 93: 445–458. To develop appropriate beneficial management practices (BMPs) for a watershed, it is essential to quantify the nutrients lost from agricultural fields and to identify the mechanisms of nutrient transport. To determine appropriate BMPs for a watershed in southeastern Saskatchewan, nutrient concentrations were measured in spring 2010 in snowmelt runoff from fertilized annual cropland (zero till) and perennial tame pastures. The majority of nutrient loss was in dissolved form, rather than as particulates. Significantly more nitrogen (N) was lost from pastures as dissolved ammonium than from cropland, while significantly more dissolved organic N was lost from croplands. Although there were no significant differences in total phosphorus (P) loss, there were significantly higher concentrations of dissolved reactive P in runoff from cropland, and significantly higher particulate P in runoff from pastures. Total carbon (C) in runoff was higher from cropland, due mainly to significantly higher dissolved organic C concentrations. Runoff from cropland contained significantly higher concentrations of dissolved potassium and sulfur, reflecting the fertilization of cropland fields with these nutrients. These preliminary results demonstrate that nutrients may be transported from agricultural lands by different mechanisms (e.g., in dissolved versus particulate forms) as a function of cropping system, requiring the development of specific types of BMPs to best control nutrient losses.

Description: Growth responses to pruning or thinning are well documented but their interactions are not, even though they are sometimes performed simultaneously. Growth responses to thinning and pruning were examined in nine plantation silvicultural experiments at five sites in southeastern mainland Australia. The species studied were Eucalyptus globulus Labill., Eucalyptus nitens (Deane and Maiden) Maiden, and Eucalyptus grandis Hill ex Maiden. Thinning from about 1100–1300 trees·ha–1 to about 300 or 500 trees·ha–1 at either age 3–4 years or 7–10 years increased the volume of sawlog crop trees in all species. Multiple lift pruning to 6.5 m height on the sawlog crop trees that retained at least 70% of the live crown length at any lift significantly reduced tree growth at only one of the six site–species combinations where both thinning and pruning were studied. And here, thinning interacted with pruning such that the pruning effects were not significant in unthinned stands because only shaded and inefficient foliage was removed. This study shows that thinning and pruning can interact to influence sawlog crop tree growth and this interaction is influenced by site.

Description: Forest stand diameter distributions adopt a wide variety of shapes, not all of which can be easily matched by specific functional forms. Sometimes, diameter distributions within a single population can be too variable to be matched by a single parametric family. Motivated by this observation, we investigate the functional regression tree (FRT) method, a recursive partitioning procedure that can be used for modelling probability density functions. We use FRT to estimate forest stand diameter distributions without needing to make assumptions about the functional form. We use an example study to demonstrate the effectiveness of the approach in which we compare the FRT method with a parameter prediction and percentile method with favourable results. We show the FRT approach to be suitable for diameter distributions that exhibit multimodality and excess skewness, which are not easily dealt with in a parametric context.